On-highway and off-highway machines typically comprise a plurality of components that cooperate to perform a variety of tasks. Failure of one or more components of the machine can often lead to a loss in some functionality of the machine, which may limit the performance capabilities of the machine. Some component failure events may be relatively minor and simple to detect and resolve, with no real collateral damage to other components of the machine. Other failure events, however, may be more serious, potentially damaging other components of the machine. For example, in hydraulic systems, a catastrophic pump failure may expel failure debris into the hydraulic fluid. This debris may be absorbed by one or more other components of the hydraulic system, potentially damaging these components.
While failure debris may be prevented from circulating through the entire hydraulic system through the use of one or more hydraulic filters, the buildup of debris may reduce the flow rate of the fluid. In hydraulic systems that deliver for performing tasks associated with the hydraulic system, the reduction in flow rate may prevent the requisite amount of fluid from reaching one or more critical components. Over time, these components may overheat, increasing the likelihood for premature wear and, eventually, failure of these components.
One method for identifying buildup of debris in hydraulic system involves the use of a pressure differential switch that monitors the input and output pressure of the filter. If the pressure differential exceeds a threshold value, the switch is triggered and a warning signal is generated notifying the machine operator that the filter may be clogged with debris, which may be indicative of a catastrophic failure of one or more components associated with the hydraulic system.
One problem with threshold-based failure detection circuits is that, because component failure often manifests itself very quickly, the pressure differential may not be large enough to trigger the alarm until after the failure event has occurred. Such late detection of failure events may potentially cause damage to other components in the fluid channel, as contaminate particles that may have been expelled by the failed component may cause damage to components located between the failed component and the fluid filter. These particles may be difficult to remove once introduced to the system, and in addition to causing catastrophic component failure to one or more components in the short-term, may lead to long-term contingent component wear as these residual particles may remain in the system indefinitely.
Furthermore, late detection of failure events for certain “critical” components such as, for example, a hydraulic pump used to drive the tracks of a machine, may dramatically decrease worksite productivity. For example, if the failure event immobilizes or otherwise disables the machine, performance of tasks that depend upon the operation of the machine may be delayed. In addition, if the machine breaks down while operating in the worksite, the machine may present an obstacle to other machines. Thus, in order to limit the effects of catastrophic component failure and/or contingent component failure on the operations of a machine and worksite, a system for early detection of machine component failure may be required.
At least one system has been developed for predicting a remaining lifespan of a hydraulic pump. For example, U.S. Pat. No. 7,082,758 (“the '758 patent) to Kageyama et al. describes a system for monitoring long-term and short-term trends associated with pressure differential data measured across a hydraulic filter. The system may predict a pump failure or estimate a pump lifespan based on the degree of deviation between the long-term trend data and the short-term trend data.
Although the system of '758 patent may be effective in predicting component failure in certain situations, the system of the '758 patent may have several disadvantages. For example, differential pressure across the pump filter will typically only begin increasing after the filter chamber has filled, which may not occur until after significant failure contaminates have already been introduced into the fluid. Consequently, by the time the system of the '758 patent detects an upward trend in the differential pressure across the filter, one or more components may have already sustained significant damage.
The presently disclosed systems and methods for early detection of machine component failure are directed toward overcoming one or more of the problems set forth above.